Field of the Invention
The present invention relates to an information processing apparatus, an information processing method, and a storage medium, and, particularly, to an information processing apparatus, an information processing method, and a storage medium for performing image processing for an image downloaded from a server accessible by a third party.
Description of the Related Art
Along with proliferation of digital cameras and smartphones, a new day has come where everyone can easily enjoy photography. When printing a digital image shot by a digital device using a printer, image processing is sometimes performed to attain a fine-looking print product. For example, if overexposure/underexposure has occurred upon shooting a digital image, it is desirable to perform appropriate image processing to compensate for this. If the eyes of a target person are in a so-called red eye state because of electronic flash upon shooting, correction is desirably performed. If a human face is included in a digital image, it is desirable to perform appropriate image processing to attain a good-looking skin color in a print product. In addition, these image processes are desirably automatically executed, because manually performing them one by one consumes much time and effort.
Generally, when automatically executing image processing, it is necessary to analyze an image or to evaluate the attribute of an image from so-called Exif tag information associated with the image data so as to perform processing suitable for the image. By the analysis and evaluation, image processing appropriate for the image is determined.
Analyzing attribute information associated with an image or image data will be referred to as image comprehension in the following description. In Japanese Patent Laid-Open No. 2010-251999, image comprehension is performed via analysis of an input image. The feature amount of the image is obtained from the result of image comprehension. Scene determination and reliability calculation of the scene are performed based on the obtained feature amount, thereby automatically executing correction processing. In this way, the result of image comprehension is mainly used to generate, to obtain, and to evaluate a feature amount.
However, image comprehension by image data itself is limited. In the method disclosed in Japanese Patent Laid-Open No. 2010-251999, a determination error may occur depending on an input image. For example, if a night scene image is erroneously determined as underexposure, the night scene image is corrected to be bright. Reversely, if an image of underexposure is erroneously determined to be a night scene image, a portion that should be corrected to be bright is not corrected as expected. As another example, assume that an image includes a person, and appropriate color balance correction needs to be performed for the person's skin color. In this case, it may be difficult to estimate the appropriate skin color of the target person based on only information obtained from the image.
When automatically performing image processing, the same problems as described above may arise, even in automatic layout generation processing. An evaluation value used to determine a layout generation condition is obtained using the analysis result of a target photo. As a result, layout generation is performed based on a feature amount that expresses image quality of an image file itself. However, it may be difficult to determine a desirable image from a feature amount obtained from image comprehension. For example, it is difficult to automatically select a desirable image that is blurred to some extent, but expressive. In addition, enormous works are necessary for one user to manually input evaluation of an image.